Skip to main content

Content Generation Service - User Guide

Quick Start​

Generate professional content in 3 steps:

from packages.content_generation import BlogPostGenerator, ContentGenerationRequest
from packages.content_generation.models import ContentType, AudienceSegment, Tone

# 1. Create generator
generator = BlogPostGenerator()

# 2. Create request
request = ContentGenerationRequest(
    content_type=ContentType.BLOG_POST,
    topic="Your Topic Here",
    audience_segment=AudienceSegment.BUSINESS_USERS,
    tone=Tone.PROFESSIONAL,
    created_by="your_user_id"
)

# 3. Generate content
content = await generator.generate_blog_post(request)
print(content.title)

Content Types​

Marketing Content​

Blog Posts:

from packages.content_generation import BlogPostGenerator

generator = BlogPostGenerator()
blog = await generator.generate_blog_post(request, seo_focus="main keyword")

# Get SEO metadata
print(f"SEO Score: {blog.quality_scores.seo_score}")
print(f"Meta: {blog.metadata['meta_description']}")
print(f"Tags: {blog.metadata['tags']}")

Email Campaigns:

from packages.content_generation import EmailCampaignGenerator

generator = EmailCampaignGenerator()
email = await generator.generate_email_campaign(
    request,
    email_type="promotional",  # or: transactional, newsletter, nurture, announcement
    generate_variants=3  # Get 3 subject line variants
)

# Get A/B test variants
for i, subject in enumerate(email.metadata['subject_lines'], 1):
    print(f"Subject {i}: {subject}")

Social Media:

from packages.content_generation import SocialMediaGenerator

generator = SocialMediaGenerator()

# Single platform
linkedin = await generator.generate_social_post(request, platform="linkedin")

# All platforms at once
posts = await generator.generate_multi_platform(request)
for platform, content in posts.items():
    print(f"{platform}: {content.character_count} chars")

Sales Content​

Sales Outreach:

from packages.content_generation import SalesOutreachGenerator

generator = SalesOutreachGenerator()

# Single outreach email
outreach = await generator.generate_sales_outreach(
    request,
    prospect_data={
        "prospect_name": "John Smith",
        "company": "Acme Corp",
        "pain_point": "manual processes"
    }
)

# Multi-touch sequence
sequence = await generator.generate_sales_sequence(request, sequence_length=5)
# Returns 5 emails with evolving tone and CTAs

Sales Proposals:

from packages.content_generation import ProposalGenerator

generator = ProposalGenerator()
proposal = await generator.generate_proposal(
    request,
    client_data={
        "client_name": "Acme Corp",
        "budget": "$50,000",
        "timeline": "3 months"
    }
)

Quality Analysis​

Check Content Quality​

from packages.content_generation import QualityScorer

scorer = QualityScorer()
scores = await scorer.score_content(
    title="My Article",
    body="Content here...",
    target_keywords=["keyword1", "keyword2"]
)

print(f"Readability: {scores['readability']['flesch_reading_ease']}")
print(f"Grammar: {scores['grammar']['grammar_score']}")
print(f"SEO: {scores['seo']['seo_score']}")
print(f"Safety: {scores['toxicity']['safety_score']}")

Check Compliance​

from packages.content_generation import ContentComplianceChecker

checker = ContentComplianceChecker()
compliance = await checker.check_compliance(
    title="Article Title",
    body="Content here...",
    requirements={
        "min_words": 500,
        "max_words": 1500
    }
)

if not compliance.passed:
    print(f"Issues: {compliance.issues}")
    print(f"Warnings: {compliance.warnings}")

A/B Testing​

Generate Variants​

from packages.content_generation import ContentVariantGenerator

variant_gen = ContentVariantGenerator()

# Tone variants
tone_variants = await variant_gen.generate_tone_variants(request)

# Audience variants
audience_variants = await variant_gen.generate_audience_variants(request)

# Compare and get best
comparison = variant_gen.compare_variants(tone_variants)
print(f"Best variant: {comparison['recommended_variant']}")

API Usage​

Generate Content​

curl -X POST http://localhost:8000/api/v1/content/generate \
  -H "Content-Type: application/json" \
  -d '{
    "content_type": "blog_post",
    "topic": "AI in Healthcare",
    "audience_segment": "business_users",
    "tone": "professional",
    "target_keywords": ["AI", "healthcare"],
    "created_by": "user_123"
  }'

Analyze Quality​

curl -X POST http://localhost:8000/api/v1/content/quality/analyze \
  -H "Content-Type: application/json" \
  -d '{
    "content": "Your content here...",
    "title": "Article Title",
    "target_keywords": ["keyword1", "keyword2"]
  }'

Best Practices​

Content Generation​

  1. Be Specific: Detailed topics get better results
  2. Use Keywords: Improves SEO and relevance
  3. Set Word Counts: Get content of appropriate length
  4. Choose Right Tone: Match your audience
  5. Enable RAG: Better context = better content

Quality Optimization​

  1. Check Readability: Aim for 60-70 Flesch score
  2. Review Grammar: Fix flagged errors
  3. Optimize SEO: Use extracted keywords
  4. Validate Safety: Ensure toxicity < 0.3
  5. Iterate: Use A/B testing to improve

Sales Content​

  1. Personalize: Use prospect data
  2. Be Concise: <250 words for outreach
  3. Multiple Touches: Use sequences (5 emails)
  4. Soft CTAs: Low-friction asks
  5. Add Value: Include resources, insights

Troubleshooting​

Low Quality Scores:

  • Check if quality libraries are installed
  • Verify content meets minimum standards
  • Review generated content manually

Slow Generation:

  • Reduce max_words
  • Disable unnecessary features
  • Check API rate limits

High Costs:

  • Set max_tokens limits
  • Use caching
  • Monitor daily budgets

Support​

  • Documentation: See full docs in docs/docs/services/personalized-content-generation/
  • Examples: Check examples/content_generation_demo.py
  • Tests: Review test files for usage patterns

User Guide Version: 1.0
Last Updated: October 9, 2025